The New York Times writes an editorial on Net Neutrality, thereby launching the issue into the mainstream. In typical Left leaning fashion, the NYT comes out for government regulation of ISPs, and argues that consumers should be guaranteed service regardless of application.
Net neutrality is about big media against big telecom, not big telecom against the consumer, though that is not what the NYT and digital elitists would like you to believe. The reality is that different types of internet consumers exist, and they should pay different amounts for different types of service. Just as the Tragedy of the Commons parable shows:
Benefits of exploitation accrue to individuals, while the costs of exploitation are distributed between all those exploiting the resource.
The telcos have done an awful job of arguing for net neutrality, and have caused such an uproar that Congress cannot back away without extracting a pound of flesh. One of the more sensible things Verizon (VZ) or AT&T (T”) could have presented at the congressional hearings is some recent insightful data ( .pdf link ) on internet use in Japan, where high-bandwidth services are most deployed.
The presentation clearly shows the following has happened in Japan once bandwidth speeds are increased:
A closer look at the data clearly shows that internet consumers already fall into two tiers.
Note this is a logarithmic chart, the consumers in the upper right are orders of magnitude above the main cluster. The model that ISP’s use ( x mb/s down, y mb/s up, always on, regardless of whether you use it or not) is clearly being exploited by some.
The American carriers must see exactly the same patterns and they are frightened. Big telco comes from a background where every single call, every bit of data must be delivered. Quality of service is paramount, but to keep this promise to all users they will need to increase backbone capacity.
The carriers are seeking a way to recover additional costs from consumers who make the decision to behave in a certain way by downloading very large amounts of media (Video, Audio, software). As peer-to-peer goes legit through paid media downloading services, and more people start taking on usage characteristics like the folks in the upper right-hand corner of the picture, someone will need to pay for more connectivity.
The other option is that one class of users is required to subsidize the usage of a smaller class – i.e. currently 96% of the users pay for the usage of 4%. The result? The price of broadband goes up, and fewer people subscribe – an outcome everyone would agree is not desirable.
In short, a small fraction of users engaging in peer to peer networking is dictating the capital expenditure requirements for backbone bandwidth. If net neutrality becomes law here in the US, carriers will be forced to pass these capital expenditure costs along to all users, regardless if they are a member of this 4% class.
Why doesn’t it surprise me that the traditional political boundaries (and newspaper editorial pages) frame the issue perfectly?
This isn’t an issue about protecting the rights of consumers. The media providers like Google (GOOG) and Yahoo (YHOO) are dead-set against this as it is money out of their pocket. They would rather have the consumers who don’t use their services pay for those who do.
Tiered usage already exists so why is it wrong to charge for it?
Other related posts on this popular topic:
You can also browse the sidebar related posts, or use the livesearch box at the upper right.
Hat Tip: .pdf link courtesy of MuniWireless
I think net neutrality is more related to QoS than bandwidth. Having large consumers pay more can still be ‘net neutral.’ Forcing Vonage subscribers to pay for a special ‘low latency’ plan is not.
Someone brought this up over at Techdirt. It’s a common question.
It’s not about QoS. It’s about charging more for using certain applications, Applications that have bandwidth profiles that break the statistical model used to build the internet.
They break the statistical model in such a way that QoS cannot be guaranteed.
In order to accomodate and adapt to these new traffic profiles more investment will be needed.
Bottom line is _someone_ needs to invest in the transport network. The telcos are not going to do it for free. You take away variable pricing flexibility and broadband prices will go up.
Having worked for several businesses with prominent web presences I’d like to say that businesses already pay for a business type connection and QoS. In some cases a business will pay 10 times what a “consumer” may be charged for the same speed of internet access. I understand and accept that a business needs to be charged as such for that sort of access but only from their ISP and not from every backbone connection that it may pass through which is what the Big Telcos are proposing. The problem with the new tiered internet proposals from companies like Verison and AT&T is that each Telco company that touches your data could conceivable charge you to transfer that data across their network or it will be put into a lower classification. If you have ever done a traceroute and watched all of the different Telcos hand off your data you’d be amazed. Now with what is being proposed any one of those can bottleneck your connection if you fail to pay them, so that OC3 connection you are pay lots of cash for from your local ISP you have to pay for again to every person downstream of you ISP. That is the root of the problem of these proposals. Companies like Google and Yahoo already pay for their connection and so do the users pulling the data from them; both ends of the connection between them are already paid for. Now apparently everyone in between each of the ends needs to paid off as well.
“In short, a small fraction of users engaging in peer to peer networking is dictating the capital expenditure requirements for backbone bandwidth.”
I don’t see the argument, those heavy *users* are customers of the telcos like Verizon, not outside companies. If the Telcos charged more for those (it’s) users then there wouldn’t be a problem. What Verizon’s CEO seems to want, is to charge other ISP’s customers extra money for traffic initiated by Verizons own users.
So
1. They could charge differential pricing if they want to, and target those exact heavy-user customers since they know who they are.
2. They could throttle the rate of those customers if they wished (which is presumably what they will do if they don’t get paid).
It’s all very stupid from Verizons side if they think Google, Yahoo MSN etc will pay. They won’t, if they ever did, they’d have to pass on the charge to Verizon’s own user base. Or the other ISP will pass the charge back via their peering agreements.
(Or do you imagine they’d subsidise the charge from all their user base?)
i.e. buy Google videos $1.99 an episode, $2.15 to Verizon users to cover the bandwidth levy.
The telcos have painted themselves into a corner by allowing everyone to believe that a 1.5Mb/s residential DSL connection and a business T1 line have the same reliability and usage profiles. You may hav paid for a 1.5Mb/s connection at home, but you didn’t pay for one that could be run 24/7. I agree the telcos and cablecos have not been forthright with what you bought (didn’t buy, you rent), but that is the reality of the situation today.
I think people are assembling many doomsday scenarios on this issue. The primary bottlenecks won’t be carrier to carrier because users who were not in the higher tier would have been policed and throttled prior to transfer.
Just wait until the peer to peer guys take down a Muni-Fi…. just like they did at universities. Peer to peer still survives at universities because much trading can be done on the intranet. The admins long throttled access to the internet for big peering nodes.
“You may hav paid for a 1.5Mb/s connection at home, but you didn’t pay for one that could be run 24/7. I agree the telcos and cablecos have not been forthright with what you bought (didn’t buy, you rent), but that is the reality of the situation today.”
There is nothing in my terms of service which state I can’t run my connection 24/7. It has vague language about being able to arbitrarily discontinue my account for any reason whatsoever, including heavy usage. In fact, there selling points are “always on” “100x faster than dial-up” or whatever. It’s not a matter of whether they’ve been “forthright” with us, it’s a matter of what they advertised and what they sold. (or rented, if you like)
Their statistical model allows for the heavy users. They are making the heavy users argument because it is a good mask for the real issue: more control and pricing flexibility.
They will continue to make hoards of money as they expand, but they want to be able to make hoards of money on top of their hoards. Also, if they can establish a tiered service, or worse, metered bandwidth, they can effectively control traffic through pricing, which opens up whole new frontiers of income possibilities. Of course, none of those would help the general public and it would muffle innovation. They must wet themselves with glee at the thought of it.
Jane
Maybe there is something in your water. Maybe the phone company put it there.
A really smart guy who knows the telcos well once told me the following: “People like to dream up all kinds of complex, sneaky scenarios that the phone companies perpetrate. The reality is most of them are so slow and inept they are incapable of any sort of devious behavior.”
Andrew
Glad you’re looking at some real data, although using Japanese fiber data to guess the future of U.S. DSL is stretching. So as a reporter I’ve recently written to the senior folks at the bells trying to get the data on traffic.
Everything I can find in the U.S., however, suggests we aren’t seeing this increase, and if we saw even a large increase – 200% in traffic – the cost of carrying it reasonably reliably is so small it shouldn’t be a policy question. (fraction of the marketing budget, for example.)
This is empirical data that should be available. I think you should make sure you you have the facts on this to back your opinion on net neutrality. The data I have suggests your argument, from first principles, doesn’t hold up.
db
The physical medium is not important – it’s the speed. ADSL that is upgraded to VDSL, or FiOS, or further out – DOCSIS 3.0 will make downloading media realistic.
The graph is from Japanese fiber users _only_. If you look at the DSL version of this graph it does not have the same pronounced cluster in the upper right.
RHK did some interesting traffic analysis in the 2000-2001 timeline and provided statistical data services on traffic volume and application usage. I wonder if they still do.
If the telcos are worried about heavy users, they can start charging per byte transferred. This is a neutral solution, it avoids subsidization, and it generates revenue that can be used for infrastructure upgrades. Customers will hate it, but they won’t like totally clogged networks either.
…from another forum, but this is a good point. Think about it…
Why stop at 2 tiers?
Why dont we just make it an 8, 10, or even 1000 tier network?
We can have one tier for web browsing. One tier for P2P. One tier for video casting. One for actual IPTV. One for IP Phones. One for multi player games, which we can then divide up per game. One for streaming music. One for FTP. One for news groups. One for Windows Updates. One for linux distros. One tier for porn, or maybe just create a whole new “internet” for porn. We can create new internet protocols for every program out there and do away with that pesky TCP/IP along with the simplicity and freedom of packet passing it provides. Then within those tiers we can tier the providers based on how much they want to pay. The backbone providers can then auction off the priority levels of each tier since they interconnect the world. Then the broadband providers can get in a spat with the backbone providers over their “cut” since they connect the consumers. And backbone providers can then start tiering broadband providers. Why should UUNet let SBC/AT&T just use their network for free? They should have to pay UUNet tiering fees to help them upgrade their backbone to support the additional subscribers they are signing up with broadband. If they aren’t willing to pay, then they should get lower priority then someone like Comcast that is willing to play.
Point of this is simple. Every single piece of the internet puzzle fits together and without one you do not have the others. You have the backbone providers, the broadband providers, the content providers, and the consumers. ALL of them are just as vital to the survival of the internet and one another. ALL of them fuel the bottom line of the others and are the very reason of their existence (stay on internet here). Content providers attract the consumers. Broadband providers provide the experience. Backbone providers interconnect the broadband providers to the content providers enabling the consumers to ultimately get the information they wanted seamlessly. Every piece of the chain here is interconnected. Every piece of the chain pays their share to survive and maintain the internet. ANYONE saying they don’t is simply an idiot. If they didn’t then we wouldnt be as far into the internet as we are today as compared to 10 years ago. We didnt get here today because free bandwidth rides were given or because providers (backbone or broadband) were just pulling bandwidth out of their rears. As the need arised and the additional cost were justified, the networks were upgraded. This happens on every end from the content providers all the way to the consumer.
I personally have not seen any content provider requesting higher priority in their packets. I have not seen a single service on the internet require higher priority that was not worked out by the developer. That is the great thing about the internet. One creates a use for it, that use is modified and somewhat perfected over time based on what they have to work with.
Mr. Schmitt, no one is saying the ISP (and lets be honest we are not talking about ISPs, we are talking about Verizon, AT&T & BellSouht, no other provider has raised this two-tiered Internet scheme) don’t have the right to charge users for what they use. Perhaps AT&T should not be marketing a $12.99 all you can eat buffet if they can’t support it.
I agree, the 95% of consumers with ‘normal’ Internet usage should not have to subisidize the bandwidth hogs, but neither should content companies.
The very basic routing system of the net relies on shortest hop routes. QoS can only be provided on the ISP’s own network and not through the entire route to the customer. Even when they can provide QoS routing on their own network, they can only do so by deprioritizng packets to others. 2+2=4, no matter which way you divide it. And a packet is a packet regardless of what data it carries.
Content companies are already developing ways on their on to improve QoS. New SIP and RTP versions have greatly reduced the bandwidth required for voip and improved quality tremendously. The same can be said for video codecs. Its not just about the pipe, its about efficient technology.
If the ILECs are allowed to segregate the net, who will be the company that develops these new codecs? AT&T, I think not. Just ask them why they developed DSL in 1982 and shelfed it for 15 years.
“I think people are assembling many doomsday scenarios on this issue. The primary bottlenecks won’t be carrier to carrier because users who were not in the higher tier would have been policed and throttled prior to transfer.”
I don’t know why that would be. AT&T(SBC) and Verizon have been so consumer and competition friendly in the past.
“A really smart guy who knows the telcos well once told me the following: “People like to dream up all kinds of complex, sneaky scenarios that the phone companies perpetrate. The reality is most of them are so slow and inept they are incapable of any sort of devious behavior.—
That’s funny a really smart guy who happened to be a former president of SBC, T.O. Gravitt, wrote in his suicide note, “Watergate was a gnat compared to the Bell System”.
Do you dispute AT&T and Verizon put misleading price increases on their phone bills? Do you dispute both companies are fined in the tens to hundreds of millions annually for violating telecom regulations?
I must be taking crazy pills – how exactly will a tiered system benefit the consumer?
You’re mixing the bandwidth hog residential user problem with the telcos plan to charge Google to deliver their high-bandwidth applications. The answer to the 4%/75% problem in your ‘Japan-fiber’ example is to charge the bandwidth hogs based on consumption. Argument foiled. Google/Yahoo/etc are already paying the telcos huge $$$ for their bandwidth.
Further, network backbone expansion has been funded and funded and funded through misleading charges on our phone bills for the last 50 years.
If you’re AOL Time-Warner, then simply throttle your consumer bandwidth to match your pricing model. Then if you want to add a pricier “Ultra Geek Plan” with more up or down bandwidth you can add more total bandwidth to the consumer network OR lower the throttle of the existing consumers a VERY small amount and thereby create the space needed for the premium service.
Obviously, consumers wouldn’t be happy about this but then it wouldn’t be a line item in a new RoadRunner TV commericial now would it :). This also addresses heavy users. Don’t slap their wrists, don’t cancel their accounts, don’t assess them penalties – just throttle them to the same speed as everyone else.
Eventually, as people use the Internet more for more everyday things and those things become more bandwidth intensive more bandwidth will need to be added. This is unavoidable.
If we were talking about electricity or space on the interstate highway system the concepts that VZ and others were proposing would be laughable. In many respects, the Internet can be viewed as a legitimate necessity for many many people. We work, play, love, and buy using the Internet. Applying models that have proven unsustainable or downright inappropriate in other industries is nothing short of chronic tunnel-vision. It’s time these companies paid a visit to the eye-doctor.
-SMT
This is not about traffic hogs, the reasoning behind this debate is that Telcos built and own the _expensive infrastructure_ therefore should be allowed to give preference to traffic generated by their customers. In other words, they are planning to sell services like Voice Over IP or Video Streaming and give preference to this premium traffic.
For us that are not premium subscribers means that our VOIP connections will be worse quality and video streaming much slower. Or worse, they may decide to completely block traffic on VOIP or related network ports unless the subscriber pays a premium.
If you think this is not already affecting you, think twice. There is a trend in the US to make the consumer pay twice for the same service. It’s like calling a mobile phone from another mobile phone, both parties are paying for the call, unlike in Europe. If I send an SMS to my wife I end up paying twice for the same message, I am taxed as the sender and again on the receiving end. Never mind that both numbers are on the same family plan under my name.
If this trend continues it won’t be long until you’ll pay iTunes for the song about to download and your ISP for the right to download it. Never mind you are already paying $49.99 per month for your _unlimited_ Internet connection. Or worse, we may soon pay for each email sent and received. Imagine sending an email to yourself, no more of those _bandwidth hogs_.
This is where the US government should intervene and set some rules to level the plain field like it did decades ago with the utilities. After all, much of this expensive Telco infrastructure was financed by US taxpayers, so again we’ll end up paying twice.
Take a look at what regulation has brought to the utilities. Stagnation. Can you think of any innovation at all in the utility sector?
There is simply no way the government can manage to anticipate and stay ahead of the curve with respect to regulating broadband.
Also – with respect to toll roads and the analogy there -perhaps you should take a look at what mark cuban wrote on his blog.
http://www.blogmaverick.com/entry/1234000267073488/
Take a look at what de-regulation did to utilities – can you say… Enron??
Yes, Andrew. Thinking that bandwidth providers want more money and control is just wacky, conspiratorial thinking. I’m wearing my tinfoil hat right now so they don’t catch on to me. Look, I don’t underestimate their ineptitude, but they’re profit driven. Being such, I imagine there are a few people looking at the long term, figuring out how they might make as much money as possible.
As far as this last post, what innovation would you like to see in the utility sector? I’m pretty pleased with always having water and electricity. On the other hand, deregulation of the energy markets resulted in price hikes, corruption and loss of service. Some things need to be regulated. Like, say, nuclear power.
J
The problem with the electricity deregulation is that lawmakers deregulated the pricing of the supply but didn’t deregulate the amount consumers could be charged. This set up the playing field for the ultimate short squeeze. Government created the environment for people to exploit.
How about allowing flexible pricing of electricity based on time of day? Or water use based on time of year? Water is a bad example because it is a finite resource, and the supply for the most part is governent controlled.
We are way off topic now.
Gentlemen:
Let us examine the foundation of logic underlying the Telecom’s advocacy by analyzing a few of the arguments made:
1. Access / usage should be metered so that those desirous of services pay for satisfying such desire;
2. Telecom’s must invest in infrastruture in order to deliver the product; Telecoms operate in a mode where “they must deliver every call, etc..”
3. “Benefits of exploitation accrue to individuals, while the costs of exploitation are distributed between all those exploiting the resource”
4. “Net neutrality is about big media against big telecom, not big telecom against the consumer, though that is not what the NYT and digital elitists would like you to believe.”
Responsive Deconstruction:
1.1) Consumers currently have available to them metered plans; Those who use the Internet little may take advantage of them. These plans are metered based upon the factors of [bandwidth] x [amount of time on-line]. AOL is an example. No one is “forced” to subsidize a “high-usage” segment of consumers
2.1) Telecoms offer no SLAs for either voice phone calls or data transfer; Indeed, we may look no further than 9-11 for an example of this:
How many people attempted to initiate both land- and cellular- based phone calls, only to be told “All Circuits Are Busy”?
Claiming that Telecoms must “deliver every call, every byte of data” is, at best a fallacy and, at worst, an outright attempt to mislead the audience.
2.2) Telecoms (and Cable operators) operate in a model in which they resell more “network” (read: capacity) than they actually posess. There are no SLAs for consumers (or content providers reaching consumers) as they would have you believe. In the event a Telecom does not wish to invest further in back-bone infrastructure, that is its choice. Another Telecom will take its place.
3.1) The word “exploitation” is pre-loaded with a negative connotation. It is used to paint “high usage” consumers into a category of being “exploiters” and taking advantage of a given situation. As the Telecom industry would find it difficult to paint itself as a “victim” of such “exploitation” the author chooses to make a victim out of those who use the network less.
If the “many” are suffering at the hands of the “few” than I propose that the Telecoms themselves are the enabling factor. As they are so wont to display usage statistics and usage patterns, once must ask: Why do they offer unlimited access plans at all? Why not offer “tiered” plans based on [bandwidth] x [amount of data sent/received]? They already have this right under law; Indeed, they do so with content providers. This also gives the lie to their supposed desire to “go after” content providers – they already do this.
4.1) “Big Media” – this is perhaps my favorite part of their premise for advocating the demise of Network Neutrality.
Big Media; IE: MSN, AOL, Yahoo!, Google, et all:
These companies *pay* for the bandwidth provided to their back-end servers. What angers AT&T is that these payments may not neccessarily go directly to AT&T but instead to AT&T’s competitors – who share infrastructure access (network neutrality); The transfer of data from Google may go through a competing Telecom and wind up on AT&T’s network as the end-consumer may be an AT&T customer, or, the customer of someone who is an AT&T customer.
Rather than competing for hosting / content provider customers, the Telecoms see an excellent opportunity to:
a) Raise rates to a point at which comopetitors customers must switch to the Telecom and;
b) Under the guise of “helping” consumers, charge a per-diem on network “transactions”
[Hypothesis / Food for Thought]
Imagine, if you will, privatization of the nation’s Interstates & Highways. Would this lead to reduced costs? Or, would a private company in direct control of a vital resource, take advantage of its position?
In closing, I propose that we, as Americans, have two choices:
1. We can allow monopolistic control of national infrastructure resources and rely upon the goodness of the corporation controlling them, or;
2. We can ensure certain basic tenets of access by establishing sound regulatory oversight of those entities exercising such control;
PS: If the “investment in infrastructure” which the Teleco’s bemoan as being too costly for too little return was not profitable than why do they continue? And why do they remain profitable under such harsh circumstances?
Johnathan I. Hyler
Trackbacks / Pingbacks